Brian L. Odlaug

NOTE TO READERS: This article was originally presented as an independent educational activity under the direction of CME LLC and published in the September 2007 issue of Psychiatric Times (2007;24[10]64-69). The ability to receive CME credits has expired. The article is presented here for your reference.


	After reading this article, you will be familiar with:

The various disorders that make up impulse control disorders (ICDs).

Pharmacological treatment options for different ICDs.

Impulse control disorders (ICDs) are common psychiatric conditions in which affected individuals typically report significant impairment in social and occupational functioning, and may incur legal and financial difficulties as well. Despite evidence of ICDs being fairly common, they remain poorly understood by the general public, clinicians, and persons with the disorders. Pharmacotherapy studies, although limited, have demonstrated that some ICDs respond well to treatment; however, there has been either very limited or, for some ICDs, no research into potential treatments. In addition, further research is needed to substantiate many of the studies that have been conducted.

Formal ICDs include pathological gambling (PG), kleptomania, trichotillomania (TTM), intermittent explosive disorder (IED), and pyromania; these disorders are characterized by difficulties in resisting urges to engage in behaviors that are excessive and/or ultimately harmful to oneself or others.

 Diagnostic criteria have also been proposed for other disorders categorized as ICDs not otherwise specified (NOS) in 

 pathological skin picking (PSP), compulsive sexual behavior (CSB), and compulsive buying (CB). ICDs are relatively common among adolescents and adults, carry significant morbidity and mortality, and can be effectively treated with behavioral and pharmacological therapies. The purpose of this review is to provide a clinical picture of these ICDs, including co-occurring psychiatric conditions (

), and to review the evidence for the pharmacological treatment of these disorders (

Core characteristics of impulse control disorders

Although the extent to which ICDs share clinical, genetic, phenomenological, and biological features is not completely understood, many ICDs share core qualities: (1) repetitive engagement in a behavior despite adverse consequences; (2) diminished control over the problematic behavior; (3) an appetitive urge or craving state prior to engagement in the problematic behavior; and (4) a hedonic quality experienced during the performance of the problematic behavior.

 These features have led to a description of ICDs as behavioral addictions.

ICDs also appear to have some clinical overlap with compulsive behaviors although this relationship is not yet completely understood. The domains of impulsivity (defined as a predisposition toward rapid, unplanned reactions to either internal or external stimuli without regard for negative consequences)

and compulsivity (defined as the performance of repetitive behaviors with the goal of reducing or preventing anxiety or distress, not to provide pleasure or gratification)

 have been considered by some as lying at opposite ends of a spectrum. Compulsivity and impulsivity may, however, occur simultaneously in a disorder or at different times within a disorder, thereby complicating both our understanding and treatment of certain behaviors.

PG is characterized by persistent and recurrent maladaptive patterns of gambling behavior and has been described as a chronic, relapsing condition. PG affects an estimated 0.9% to 1.6% of persons in the United States.

 Men tend to have higher rates of PG and start gambling at an earlier age than women.

 Women, who represent approximately 32% of pathological gamblers in the United States, appear to progress to problematic gambling faster than men.

 PG is associated with impaired functioning; reduced quality of life; and high rates of bankruptcy, divorce, and incarceration. Financial and marital problems are common.

 Many pathological gamblers engage in illegal behavior, such as stealing, embezzlement, and writing bad checks to fund their gambling or to attempt to fix past gambling losses.

 Suicide attempts have been reported in 17% of individuals in treatment for PG.

Kleptomania is characterized by repetitive, uncontrollable stealing of items not needed for personal use.

 Although kleptomania typically has its onset in late adolescence or early adulthood,

 the disorder has been reported in children as young as 4 years

 Intense guilt and shame are commonly reported by those with kleptomania. Stolen items are typically hoarded, given away, returned to the store, or thrown away.

 Many individuals with kleptomania (64% to 87%) have been apprehended at some time as a result of their stealing behavior.

TTM is characterized by repetitive, intentional hair pulling that causes noticeable hair loss and results in clinically significant levels of distress or functional impairment.

 Although trichotillomania appears to be relatively common, with an estimated prevalence between 1% and 3%,

 only 65% of those with TTM have sought treatment for their hair pulling.

 Significant social and occupational disability is common, with 34.6% of individuals reporting daily interference with job duties and 47% reporting avoidance of social situations, such as dating or participating in group activities.

IED is characterized by recurrent, significant outbursts of aggression, often leading to assaults against people or property, which are disproportionate to outside stressors and not better explained by another psychiatric diagnosis.

 Individuals suffering from IED regard their behavior as distressing and problematic.

 Outbursts are generally short-lived (usually less than 30 minutes) and frequent (multiple times per month).

 Legal and occupational difficulties are common.

 Recent research suggests that IED may be common, with 6.3% of a community sample meeting criteria for lifetime IED.

Pyromania is characterized by the following diagnostic criteria:

Deliberate and purposeful fire setting on more than one occasion.

Tension or affective arousal before the act.

Fascination with, interest in, curiosity about, or attraction to fire and its situational contexts.

Pleasure, gratification, or relief when setting fires or when either witnessing or participating in their aftermath.

Although pyromania is considered by some to be a rare disorder, in a study of 107 patients with depression, 3 (2.8%) met 

 and a recent study of 204 psychiatric inpatients revealed that 3.4% (n = 7) met current 

 criteria for pyromania and 5.9% (n = 12) had lifetime symptoms of pyromania.

 Fire setting among individuals with pyromania often does not meet the legal definition of arson.

PSP is characterized by the repetitive or compulsive picking of skin to the point of causing tissue damage. PSP has an estimated prevalence of 4% in the collegiate population and 2% in dermatology clinic patients.

 The afflicted person frequently reports shame and embarrassment and the avoidance of social situations.

 People who engage in this behavior typically spend a significant amount of time picking, often several hours each day. Most often they pick their face, but any body part may be the focus-for example, torso, arms, hands, or legs. The picking often leads to infections and/or significant scarring.

Although CB is not specifically recognized in 

 the following diagnostic criteria have been proposed for this disorder: (1) a preoccupation with buying (characterized by either an irresistible, intrusive and/or senseless preoccupation with buying or buying more than one can afford, buying unneeded items, or shopping for a longer time than originally intended); and (2) having the preoccupation with buying result in marked distress, interfere with social or occupational functioning, and cause financial problems.

 In a recent random-sample study of 2513 US adults, 5.8% of those surveyed were positive for compulsive buying.

Purchased items often go unused, are given away, or are returned to the store. Although CB is initially pleasurable, feelings of guilt, embarrassment, and shame follow buying binges.

CSB is described as excessive or uncontrolled sexual behavior or thoughts that lead to marked distress and social, occupational, legal, and/or financial consequences.

 CSB can involve a wide range of sexual behaviors, either nonparaphilic (eg, masturbation, promiscuity, pornography) or paraphilic (eg, exhibitionism, voyeurism, fetishes), that have become excessive or uncontrolled. The behavior is usually driven by either pleasure seeking or anxiety reduction.

 The prevalence of CSB in adults is estimated to range from 3% to 6%.

Despite their high prevalence in the general population

 the pharmacological treatment of ICDs has been relatively understudied. There are no FDA-approved medications for the treatment of any ICD. Although numerous medications have been tested in open-label trials and case reports, our focus is only on findings from double-blind, placebo-controlled trials.

Because impulsive behaviors have been associated with low levels of the serotonin metabolite 5- hydroxyindole acetic acid within the cerebrospinal fluid and with blunted serotonergic response to a serotonergic stimulus (metachlorophenylpiperazine) within the prefrontal cortex (visualized using positron emission tomography),

 one hypothesis is that decreased serotonin function within the prefrontal cortex may engender disinhibition and contribute to ICD behaviors. Thus, drugs targeting serotonin neurotransmission have been examined.

A single patient with PG responded to clomipramine in a double-blind, placebo-controlled study.

 After receiving placebo for 10 weeks without response, she was treated with 125 mg of closmipramine and then reported a 90% improvement in gambling symptoms. There have been no further controlled studies of clomipramine for PG to confirm these limited results.

In a double-blind, placebo-controlled study using sertraline, 60 individuals with PG were treated for 6 months (mean dosage, 95 mg/d).

 At the end of the study, there were 23 (74%) responders in the sertraline group and 21 (72%) in the placebo group; sertraline did not demonstrate superiority over placebo.

A double-blind, 16-week, crossover study of 15 patients with PG who received fluvoxamine demonstrated a statistically significant difference compared with placebo.

 Interpretation of the study is complicated, however, by a phase order treatment interaction (ie, the medication did not separate from placebo during the first phase but did in the second phase). A 6-month, double-blind, placebo-controlled trial of 32 gamblers who received fluvoxamine failed to show statistical significance compared with placebo; however, the results of this study were complicated by high rates of treatment discontinuation (only 3 subjects receiving medication completed the study)and a high placebo response rate (59%).

An 8-week, double-blind, placebo-controlled study of paroxetine demonstrated possible efficacy in the treatment of PG.

 A larger multicenter double-blind, placebo-controlled trial, however, failed to reproduce the results.

 In this second study, a high placebo response rate was observed, with 48% of those assigned to placebo and 59% of those taking paroxetine considered responders.

An open-label, 12-week trial of escitalopram with an 8-week, double-blind discontinuation period for responders was recently undertaken in pathological gamblers with marked anxiety.

 Of the 13 patients treated with a mean dose of 25.4 mg of escitalopram daily, 62% were considered responders in terms of both PG and anxiety symptom reduction. Of the 6 patients who completed the study, the 4 who were responders were entered into an 8-week, double-blind discontinuation phase. For the 3 assigned to escitalopram, improvement continued for the next 8 weeks, whereas both gambling symptoms and anxiety returned within 4 weeks for the one patient assigned to placebo.

The findings from a 12-week, double-blind study of buproprion, the only non-SSRI antidepressant included in a double-blind study, showed that there was no statistical difference in rates of improvement between the 2 groups. In fact, 35.7% of bupropion and 47.1% of placebo recipients responded to treatment.

Although the results have been mixed, the treatment data from these trials suggest that SSRIs may be beneficial for some persons with PG. There is also some indication that those with PG and anxiety may preferentially respond to these agents, but which variables predict response to serotonergic agents has not been fully examined. No positive, randomized, placebo-controlled study of an antidepressant in PG has been successfully reproduced, and there have been no comparison studies. Given the limitations of the existing literature, it is not yet possible to conclude which antidepressants should be recommended and which individuals with PG are more likely to respond to these agents.

Several controlled trials using antidepressants have been conducted in TTM. One study compared clomipramine and desipramine in a 10-week, double-blind, crossover (5 weeks for each agent) design (following 2 weeks of single-blind placebo lead-in).

 Of 13 participants, 12 had significant improvement with clomipramine, whereas only 1 patient preferentially responded to desipramine.

Fluoxetine has been studied in 4 randomized trials with conflicting results. Fluoxetine failed to separate from placebo in one 6-week, double-blind, crossover study.

 Another study compared fluoxetine with clomipramine in a double-blind, randomized, 20-week, crossover design (10 weeks on each agent) and found that both agents demonstrated a similar positive treatment effect.

 However, a third double-blind, placebo-controlled crossover study of 16 patients failed to show significant improvement with fluoxetine compared with placebo.

 Finally, a fourth study (12 weeks) that compared fluoxetine with behavioral therapy or a wait list found that although behavioral therapy demonstrated improvement in hair pulling compared with the wait list condition, fluoxetine did not.

Clomipramine may benefit individuals with TTM, but there are conflicting data on the use of SSRIs. As in other ICDs, these results suggest that some patients with TTM may respond to these agents, but data on which individuals will respond to these agents has not been examined.

Two double-blind studies of fluoxetine have been conducted for the treatment of PSP. In the first trial, 21 patients received 10 weeks of either placebo or active medication. The fluoxetine group improved significantly more than the placebo group at a mean dosage of 55 mg/d.

 The second study consisted of an open-label component for 6 weeks followed by a double-blind, 6-week discontinuation for responders. Eight subjects responded to open-label treatment; the 4 that were randomized to the fluoxetine group maintained their improvement, while the placebo group returned to their baseline symptom severity level.

With 2 positive studies using fluoxetine in the treatment of PSP, there is some support for its use in this disorder. However, the studies consisted of small samples and were of short duration, therefore the data are limited. Given the mixed results seen for fluoxetine in the treatment of TTM, recommendations regarding the fluoxetine treatment of PSP await longer, larger studies.

Pharmacotherapy using antidepressants in treating CB has been examined in 3 double-blind, randomized, placebo-controlled trials, and 1 trial using an open-label design followed by a double-blind discontinuation. In one double-blind fluvoxamine study, 37 persons were treated for 13 weeks. Only 9 of 20 patients assigned to medication were responders (mean dosage, 215 mg/d), and this rate did not differ significantly from that in the placebo group (8 of 17 were responders).

 In the second study, Black and colleagues

 treated 23 patients for 9 weeks. With a mean dosage of 200 mg/d, no differences in response rates were observed between active drug and placebo. The third study involved a 7-week, open-label trial of citalopram with responders randomized to 9 weeks of double-blind medication or placebo.

 Individuals in the citalopram group demonstrated statistically significant decreases in the frequency of shopping and the intensity of thoughts and urges concerning shopping. Interestingly, a recent 7-week, open-label study followed by 9 weeks of randomization using escitalopram found that there were no statistical differences in relapse rates between those receiving the drug and those assigned to placebo.

There is scant evidence concerning effective treatments for CB. As in TTM, trials of SSRIs have produced mixed results. The inconsistency of these results further suggests that there may be neurobiological heterogeneity within the disorders that has not yet been fully examined.

In the only double-blind study of CSB, 28 men were assigned to 12 weeks of either citalopram or placebo. The citalopram group (mean effective dosage, 43.4 mg/d)

 demonstrated significant reductions in sexual desire, frequency of masturbation, and use of pornography.

These results suggest that serotonergic agents may be beneficial in treating CSB. Given the mixed results of these agents in other ICDs, however, the question remains whether similar inconsistent results will emerge if further studies of these agents are performed.

In an open-label study with a double-blind discontinuation phase, escitalopram was examined in 24 patients with kleptomania. There was no statistical difference in rates of relapse between those taking medication and those assigned a placebo.

 This study suggests that serotonergic agents do not appear beneficial for patients suffering from kleptomania.

Comorbidity studies suggest that individuals with ICDs may experience elevated rates of co-occurring bipolar spectrum disorders (bipolar type I, bipolar type II, and bipolar NOS).

 Similarly, those with bipolar disorder may demonstrate impulsive behaviors that resemble the symptoms of ICDs. Mood stabilizers may be effective in controlling symptoms of certain ICDs. Double-blind studies of mood stabilizers in the treatment of ICDs, however, are limited.

In a double-blind, placebo-controlled study of 40 patients with PG and bipolar spectrum disorders, sustained-release lithium carbonate (mean lithium level, 0.87 mEq/L) was superior to placebo in reducing PG symptoms during 10 weeks of treatment.

 Lithium treatment resulted in statistically significant improvement in gambling thoughts and urges, but no differences were found in the amount of money lost, the number of gambling episodes per week, or the length of each gambling episode.

As one of the few studies to target a specific subtype of an ICD (ie, gamblers with bipolar disorders), this study suggests that pharmacological efficacy may depend on targeting specific ICD subtypes.

In a 12-week, randomized, double-blind study, 96 patients with cluster B personality disorders, 116 patients with IED, and 34 patients with posttraumatic stress disorder were randomized to receive divalproex sodium or placebo. Patients who met criteria for more than 1 of the 3 diagnostic categories were assigned to the diagnostic group most closely associated with their impulsive aggression. The study found that divalproex sodium had no significant influence on aggression in the participants with a primary diagnosis of IED.

This was the only study of drug treatment for IED, and it failed to demonstrate benefit. IED is a disabling disorder with significant public health impact, and treatment studies for this disorder are needed.

) have been proposed as underlying the reward seeking that triggers the release of dopamine and produces feelings of pleasure.

 Opioid antagonists are thought to decrease dopamine neurotransmission in the nucleus accumbens and linked motivational neurocircuitry, thus dampening ICD-related excitement and cravings.

A study of 45 patients who completed a 12-week, double-blind trial of naltrexone (mean dosage, 188 mg/d) for PG showed significant improvement compared with placebo.

 Naltrexone was effective in reducing the frequency and intensity of gambling urges, as well as gambling behavior. A multicenter study further demonstrated the efficacy of another opioid antagonist, nalmefene, in the treatment of PG. In a sample of 207 subjects, nalmefene demonstrated statistically significant improvement in gambling symptoms compared with placebo in a 16-week, double-blind trial.

These 2 large studies lend support to treating PG as an addiction and to using agents that have shown benefit in substance use disorders. Both studies also targeted gamblers with at least moderate urges to gamble (those with only minimal urges were not enrolled) and therefore support the notion of looking beyond the diagnosis and focusing on subtypes within the disorder.

A 6-week, placebo-controlled, randomized trial of the opioid antagonist, naltrexone, demonstrated significant improvement on 1 measure of TTM symptoms compared with placebo.

 Nearly half of those taking naltrexone (compared with none in the placebo group) reduced hair pulling by more than 50%.

As in PG, this study suggests that a subtype of ICD (for example, those with urges to engage in a behavior) may benefit from pharmacotherapy targeting those urges.

Dopaminergic systems, which influence rewarding and reinforcing behaviors, have been implicated in ICDs. Animal studies demonstrate that the key pathophysiology of addiction is mediated through glutamate and dopamine within the core of the nucleus accumbens. Repeated behaviors associated with reward produce persistent neuroplasticity in extracellular glutamate levels in the nucleus accumbens.

 The cysteine-glutamate antiporters are the source of extracellular glutamate, and they modulate the release of glutamate and dopamine.

 Recent research hypothesizes that agents that can improve glutamatergic tone within the nucleus accumbens may reduce reward-seeking behavior in persons with addictions.

-acetylcysteine (NAC), an amino acid and cysteine prodrug, appears to increase extracellular levels of glutamate and stimulate inhibitory metabotropic glutamate receptors thereby reducing synaptic release of glutamate. Restoring extracellular glutamate concentration in the nucleus accumbens appears to block reinstitution of compulsive behaviors and decrease cravings (

). Twenty seven patients with PG were treated in an 8-week, open-label trial of NAC. Responders to the open-label phase were randomized to 6 weeks of double-blind NAC or placebo. During the open-label phase, gambling symptoms were significantly reduced, with 59% of the study participants meeting responder criteria. Of those who entered the double-blind phase and were assigned to NAC, 83.3% met responder criteria at the end point, compared with only 28.6% of those assigned to placebo.

This study focused on gamblers with at least moderate urges to gamble and used an agent that specifically targeted those urges. The study suggests that NAC may be beneficial in gamblers with urges to gamble. The study further supports the examination of glutamatergic agents in PG and other ICDs as well.

ICDs have received relatively little attention from clinicians and researchers. Despite a prevalence similar to or greater than that for schizophrenia and bipolar disorder, much less research has been undertaken. Nevertheless, the treatment data presented in this article represent significant advances from only several years ago.

With few published studies that even approximate large, well-powered controlled trials, it is not possible to make treatment recommendations with a substantial degree of confidence. No drugs are currently FDA-approved for the treatment of any of the formal ICDs. Specific drug therapies offer promise for the effective treatment of PG, although PG studies are substantially limited. Most published studies have employed relatively small sample sizes, are of limited duration, and involve potentially nonrepresentative clinical groups (eg, those without co-occurring psychiatric disorders). Heterogeneity of treatment samples may also complicate identification of effective treatments. At present, questions such as which medication to use and for whom, or the optimal duration of pharmacotherapy cannot be sufficiently addressed with the available data.

For other ICDs, there are less available data to generate empirically supported treatment recommendations. In conjunction with emerging epidemiological data supporting a relatively high prevalence of ICDs, the small amount of empirical data on effective treatments for ICDs highlights the clinical need for additional research. Definitive pharmacological treatment recommendations await completion of large-scale, controlled treatment studies for these disorders and comparative investigations of pharmacological agents. Advances in these areas hold the potential for significantly improving the lives of individuals with ICDs and those directly or indirectly affected by their conditions.

Dr Grant is associate professor in the department of psychiatry, Brian Odlaug is a research assistant, and Dr Kim is professor of psychiatry at the University of Minnesota, Twin Cities.

Dr Grant reports that he has received research support from Forest Pharmaceuticals, Somaxon Pharmaceuticals, and GlaxoSmithKline. Dr Kim and Mr Odlaug report no conflicts of interest concerning the subject matter of this article.

Diagnostic and Statistical Manual of Mental Disorders

)
	. Washington, DC: American Psychiatric Publishing, Inc; 2000.
	

Pathological Gambling: A Clinical Guide to Treatment

. Washington, DC: American Psychiatric Publishing, Inc; 2004.
	

 Moeller FG, Barratt ES, Dougherty DM, et al. Psychiatric aspects of impulsivity. 

 2001;158:1783-1793.4. National Opinion Research Center at the University of Chicago. Gambling Impact and Behavior Study: Report to the National Gambling Impact Study Commission, 1999. Available at: http://www.norc.uchicago.edu/new/ gamb-fin.htm. Accessed June 7, 2007.
	

 Grant JE, Kim SW. Demographic and clinical features of 131 adult pathological gamblers. 

 Potenza MN, Steinberg MA, McLaughlin SD, et al. Illegal behaviors in problem gambling: an analysis of data from a gambling helpline. 

 Petry NM, Kiluk BD. Suicidal ideation and suicide attempts in treatment-seeking pathological gamblers. 

 McElroy SL, Pope HG, Hudson JI, et al. Kleptomania: a report of 20 cases. 

 Phelan J. Childhood kleptmania: 2 clinical case studies with implications for further research. 

 McNeilly DP, Burke WJ. Stealing lately: a case of late-onset kleptomania.

 Grant JE,Kim SW. Clinical characteristics and associated psycho-pathology of 22 patients with kleptomania. 

 Christenson GA, Pyle RL, Mitchell JE. Estimated prevalence of trichotillomania in college students. 

 Woods DW, Flessner CA, Franklin ME, et al. The trichotillomania impact project (TIP)
	: exploring phenomenology, functional impairment, and treatment utilization. 

 Keuthen NJ, Franklin ME, Bohne A, et al. Functional impairment associated with trichotillomania and implications for treatment development. Paper presented at: Trichotillomania: Psychopathology and Treatment Development Symposium, 36th annual meeting of the Association for the Advancement of Behavior Therapy; November 2002; Reno, Nev.
	

 McElroy SL, Soutullo CA, Beckman DA, et al. 

 intermittent explosive disorder: a report of 27 cases. 

 Coccaro EF, Schmidt CA, Samuels JF, et al. Lifetime and 1-month prevalence rates of intermittent explosive disorder in a community sample. 

 Lejoyeux M, Arbaretaz M, McLoughlin M, Ades J. Impulse control disorders and depression. 

 Grant JE, Levine L, Kim D, Potenza MN. Impulse control disorders in adult psychiatric inpatients. 

 Grant JE. Pyromania: clinical characteristics and psychiatric comorbidity. 

 Keuthen NJ, Deckersbach T, Wilhelm S, et al. Repetitive skin-picking in a student population and comparison with a sample of self-injurious skin-pickers. 

 Griesemer RD. Emotionally triggered disease in a dermatologic practice. 

 Gupta MA, Gupta AK, Haberman HF. The self-inflicted dermatoses: a critical review. 

 Arnold LM, Auchenbach MB, McElroy SL. Psychogenic excoriation: clinical features, proposed diagnostic criteria, epidemiology and approaches to treatment. 

 McElroy SL, Keck PE, Pope HG, et al. Compulsive buying: a report of 20 cases. 

 Koran LM, Faber RJ, Aboujaoude E, et al. Estimated prevalence of compulsive buying behavior in the United States. 

 Black DW, Kehrberg LL, Flumerfelt DL, Schlosser SS. Characteristics of 36 subjects reporting compulsive sexual behavior. 

Articles

Impulse control disorders are common psychiatric conditions in which affected individuals typically report significant impairment in social and occupational functioning, and may incur legal and financial difficulties as well.

Brian L. Odlaug

Articles

Impulse Control Disorders: Clinical Characteristics and Pharmacological Management